ECE2019 Guided Posters Adrenal and Neuroendocrine - Basic (14 abstracts)
1Centre de Recherche en Cancérologie de Lyon, Inserm U1052, CNRS UMR5286, Université Lyon 1, Lyon, France; 2Centre de Pathologie Est, Groupement Hospitalier Est, Hospices Civils de Lyon, Bron, France; 3Universite Lyon 1, Service de Neurochirurgie, Hôpital Neurologique, Hospices Civils de Lyon, Bron, France; 4Fédération dEndocrinologie, Groupement Hospitalier Est, Hospices Civils de Lyon, Bron, France; 5Faculté de Médecine Lyon Est, Université Lyon 1, Lyon, France.
Introduction: Pituitary Neuroendocrine tumours (PitNETs) present heterogenic characteristics based on their hormonal expression and secretion. While most PitNETs have a slow progression rate, a subset of them exhibit an aggressive behaviour with recurrence properties despite current surgery, radio- and chemotherapy. To explore new clinical strategies based on immunotherapy we initiated a systemic cartography of the pituitary tumour immune microenvironment (PiTME).
Methods: Flow cytometry (FACS) and immunohistochemical (IHC) analysis were performed on a cohort of 43 freshly resected human PiNETs (18 Somatotroph and 25 Gonadotroph tumours). Patient derived xenografts (PDXs) were also performed using Rag2KO mice to further explore the interaction between pituitary tumour cells and mouse macrophages.
Results: Identification of various content of infiltrating T, B, NK and macrophage immune population was confirmed by FACS in all PitNETs. Interestingly, we found Gonadotroph and Somatotroph tumours to show a respective and specific immune signature. While Gonadotroph tumours were presenting a significant increase of CD68+ macrophage infiltration, Somatotroph tumours were showing a larger number of infiltrating T cells. Those observations were subsequently validated by IHC on FFPE tumour-sections, confirming the increased number of CD68+/CD163+ macrophages found within Gonadotroph lesions. Analysis of clinical data demonstrated that macrophage infiltration was correlated with local invasion of Gonadotroph-tumours. Interestingly, we also found those invasive lesions to have a significant histological reduction of collagen deposition, pointing a possible mechanism that could link invasion to a collagen-matrix remodelling by infiltrating macrophages in Gonadotroph tumours. Finally, through the use of a Rag2KO-PDX-model of PitNETs, we found mouse macrophages capable of an increase infiltration potential in grafted Gonadotroph tumours compared to Somatotroph grafted lesions, a result we further found associated with a reduced collagen deposition in Gonadotroph grafted tumours.
Conclusion: Our work underlines the specific immune landscape that respectively exist in Somatotroph and Gonadotroph tumours. It further pinpoints the potent role of CD68+/CD163+ macrophages in Gonadotroph tumour invasion through a collagen-matrix remodelling. Future experiments are needed to explore whether macrophages ablation or re-education could represent beneficial immune therapies for PitNET patients.